Packet services are carried over optical networks which operate a Layer 0 (photonic, such as Dense Wave Division Multiplexing (DWDM)) and Layer 1 (Time Division Multiplexing (TDM), such as Optical Transport Network (OTN)). Optical networks at L0/L1 provide service restoration due to faults such as mesh restoration through a control plane, Software Defined Networking (SDN) controller, Path Computation Element (PCE), etc. It is important for service providers to keep track of service availability for customer's Service Layer Agreements (SLAs). Conventionally, services are monitored via Performance Monitoring (PM) data which typically log 15 minute/24 hour data on a path terminating Connection Termination Point (CTP). Such PM data can be used to determine the aggregate number of seconds that are faulted in any 15 minute/24 hour duration. Further, such PM data is logged for historical monitoring. When there is a fault and subsequent restoration, the conventional PM data can provide insight into the duration of the fault. Also, higher layers (packet) can use this PM data for restoration action in such layers. Disadvantageously, the conventional PM data approach provides an approximation of the restoration time in tens of seconds accuracy. It is desired to provide such data at better accuracy than tens of seconds for SLA tracking, higher layer restoration initiation, and the like.